diff options
| author | dfukalov <daniil.fukalov@amd.com> | 2019-11-05 17:30:52 +0300 |
|---|---|---|
| committer | dfukalov <daniil.fukalov@amd.com> | 2019-11-06 13:55:48 +0300 |
| commit | 47a5c36b37f033496de01b48cef0b2f1903c33cc (patch) | |
| tree | 0024fa56f8b7c2f4dcf42e341d77db7d0393f026 /llvm/test/Transforms/LoopUnroll/AMDGPU | |
| parent | 59f063b89c518ae81467f6015d1c428c61583f71 (diff) | |
| download | bcm5719-llvm-47a5c36b37f033496de01b48cef0b2f1903c33cc.tar.gz bcm5719-llvm-47a5c36b37f033496de01b48cef0b2f1903c33cc.zip | |
[AMDGPU] Improve code size cost model (part 2)
Summary: Added estimations for ShuffleVector, some cast and arithmetic instructions
Reviewers: rampitec
Reviewed By: rampitec
Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, zzheng, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D69629
Diffstat (limited to 'llvm/test/Transforms/LoopUnroll/AMDGPU')
| -rw-r--r-- | llvm/test/Transforms/LoopUnroll/AMDGPU/unroll-for-private.ll | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/llvm/test/Transforms/LoopUnroll/AMDGPU/unroll-for-private.ll b/llvm/test/Transforms/LoopUnroll/AMDGPU/unroll-for-private.ll index e2606bb69cf..74e124cf6af 100644 --- a/llvm/test/Transforms/LoopUnroll/AMDGPU/unroll-for-private.ll +++ b/llvm/test/Transforms/LoopUnroll/AMDGPU/unroll-for-private.ll @@ -1,4 +1,4 @@ -; RUN: opt -data-layout=A5 -mtriple=amdgcn-unknown-amdhsa -loop-unroll -S -amdgpu-unroll-threshold-private=20000 %s | FileCheck %s +; RUN: opt -data-layout=A5 -mtriple=amdgcn-unknown-amdhsa -loop-unroll -S -amdgpu-unroll-threshold-private=12000 %s | FileCheck %s ; Check that we full unroll loop to be able to eliminate alloca ; CHECK-LABEL: @non_invariant_ind |
